Dense

Dense is the standard fully connected layer used in most typical neural networks.

Each output feature is computed as the weighted sum of all input features, followed by a bias term.

Bases: Module

A fully connected layer.

Applies the affine transformation

y = xW + b

to the input tensor.

Parameters:

Name	Type	Description	Default
in_shape	int	Number of input features.	required
out_shape	int	Number of output features.	required

Shapes

Input:
    (N, in_features)

Output:
    (N, out_features)

Example

layer = ntk.Dense(784, 128)

y = layer(x)

Example

import neuraltoolkit as ntk
import numpy as np

sample_count = 50
input_features = 10

x = ntk.Tensor(
    np.random.rand(sample_count, input_features)
)

layer = ntk.Dense(
    in_shape=input_features,
    out_shape=4
)

y = layer(x)